With the increasing development and change of the communications technologies, a radio technology gradually develops into a broadband radio network, and services are also developed from voice services to multimedia services. In addition, demands for multimedia services such as a video, a video conference, and a game are continuously growing. Generally, these services are called streaming services and are quite sensitive to quality of service (QoS, Quality of Service) of delay, which is reflected in queue time delay limitation in an allocation principle of radio resource management (RRM, Radio Resource Management). Based on different requirements for time delay, regarding energy saving, there is room for optimization of varying degrees in a design of a system transmission resource allocation algorithm.
For the design of the system transmission resource allocation algorithm, channel state information (CSI, Channel State Information) is very important. In an actual system, the CSI is always not ideal enough, and the CSI may be not accurate enough because of an accuracy limitation caused by a difference between measurement time and use time. Therefore, channel state information at transmitter (CSIT, Channel State Information at Transmitter) determined by a base station based on the CSI is also inaccurate. However, in many existing designs of the system transmission resource allocation algorithm, it is assumed that an RRM algorithm of a system is ideal, and a packet error problem caused by CSIT inaccuracy is ignored. However, inaccuracy of CSIT information will affect an actual instant channel capacity of the system, and when the system capacity cannot be determined, a packet error probability cannot approach zero even when channel code with a strong error-correcting feature is used.
In a study, the inventor found that a resource allocation method in the prior art has the following defects: The problem of non-ideal CSI is not considered in the prior art, and as a result, a relatively large packet error ratio is easily produced when a base station transmits data, which degrades system performance. In addition, in an existing resource allocation method for services sensitive to time delay, total power consumption of sending data by the base station is relatively high. Further, there is no such a method for lowering the total power consumption of the system on the basis of considering a trade-off relationship between power and time delay of the base station.